1. Field of the Invention
The present invention relates to a random number generator, and in particular to a random number generator for generating random numbers that are utilized as seeds by a random number generator to generate random numbers having a greater degree of randomness.
2. Background of the Invention
Random number generation and random number generators are typically used to generate random numbers for simulations in computational science and engineering. Random numbers are used to model the randomness of random noise or perturbations and quantum processes. Random number generators are also used in many other applications including slot machines and cryptography.
Because random number generators typically generate random numbers using deterministic algorithms, random number generators are often called pseudo-random number generators. In many applications, the quality of the overall randomness (i.e., how closely the random numbers resemble truly random sequences) of the numbers generated is not particularly important. However, in many of the areas where random number generation is heavily used, e.g., security applications and sophisticated scientific modeling applications, the quality or overall randomness of the random numbers generated is very important.
Typical random number generators rely on one or more seeds to generate random numbers. The seeds are fed to the random number generator. The random number generator processes the seeds through a series of algorithms to generate a random number. Typically, use of the same seed number or same combination of seeds will generate the same random number. As a result, the overall randomness of the seed number may be critical in determining the overall quality of the random numbers generated by a particular random number generator.
Prior art systems for enhancing the overall quality of the one or more seeds include schemes that involve measuring the random phenomenon of nuclear decay and the light of a light-emitting diode. Other known systems involve measuring the time between randomly generated analog pulse signals. Although prior art systems may provide one or more high quality seeds, prior art systems often require complex manufacturing processes, higher manufacturing costs, and are difficult to contain within a compact space such as a chip. In addition, many prior art schemes include digital signals that may be compromised by outside sources.